I work in a public sector undertaking — technical departments, very process-driven, very numbers-heavy. When people hear that, they usually assume GMAT must have been easy for me. Quant? Probably. But the exam has three sections, and I was genuinely struggling with two of them.
This is how I fixed that, cracked a 645, and ended up with an ISB admit in Round 3.
Why GMAT, Why Now
The decision to take the GMAT wasn't sudden. I'd been sitting in technical meetings for a few years, doing solid work, but I started noticing a gap. The business side of things — the complete picture of how organizations think, how decisions get made — I wasn't getting that exposure. The world is also changing fast, and I could see that clearly from my PSU
vantage point. That realization had been building for a while, but with work commitments piling up, I kept pushing the decision to "next quarter."
In 2024, I finally stopped waiting. I came across e-GMAT and the flexibility of their platform immediately stood out — recorded lectures I could attend at my own pace, not tied to any fixed schedule. That mattered a lot given my work situation. I enrolled, took my first diagnostic mock and scored low 500, and got a reality check. The score wasn't going to get me anywhere near my target schools.
Strong engineers tend to assume their quant instinct will carry them through most of the exam. I was about to spend the next few months discovering just how wrong that assumption was.
Starting from Scratch: The Foundation Stage
When I connected with the e-GMAT team in my first coach-connect session, they mapped out exactly how I needed to approach my preparation — which topics needed focus, how much time to allocate, and what sequence to follow. That kind of structured plan was something I hadn't had before. Instead of wandering through YouTube videos and random question sets, I now had a clear pathway.
What helped immediately was how the content was structured. Bite-sized video lessons — not hour-long lectures, but short, focused modules I could actually complete after work. The flow was very deliberate: concept videos first, learning activities, then practice files. Each step built on the last. I wasn't just consuming content; I was actually checking whether I'd understood it before moving ahead.
The learning level grades were a useful mirror. Getting a C on a module after I thought I understood it was sometimes annoying, but it kept me honest. There were concepts I thought I had locked down that I clearly hadn't — the grades made that visible rather than letting me discover it in a mock three weeks later.
The untimed practice in the early stages was something I was initially skeptical about. I kept thinking I should be training under time pressure from day one. But the untimed practice at the start forced me to actually focus on process — was I setting up the problem correctly, was I applying the right method? Once the method was solid, the speed followed naturally.
The PACE Engine: An Ego Check for Engineers
Quant was supposed to be my strong section. I walked into the diagnostic quizzes confident. What happened next was a lesson in intellectual humility.
Number properties broke me. I was certain I knew that topic — prime numbers, factors, divisibility, all things I'd studied in engineering entrance exams years ago. The PACE engine ran its diagnostics and pointed directly at that gap. Edge cases in factor calculation. Things I'd assumed I knew but had essentially forgotten.
The PACE framework doesn't let you skip something just because you think you know it. It makes you prove it. And when I couldn't, it redirected my preparation to where the actual gaps were rather than where I imagined them to be. I ended up saving over 25-30 hours of preparation time on topics I would have over-prepared — and that time got redirected to verbal and DI, where I actually needed it.
What also helped here was focusing on process — not just knowing formulas, but building a systematic approach to each question type. The emphasis on applying structure (what do I know, what am I solving for, what approach fits this problem type) was something I'd never done consciously before.
Verbal: The Section I Was Actually Afraid Of
Critical Reasoning clicked first, and honestly it's because the approach e-GMAT teaches resonates with how I already thought. You identify the argument's structure — what's the premise, what's the conclusion, what assumption connects them. Then you pre-think: before even reading the answer choices, form your own prediction of what a correct answer should look like. Out of five choices, two will almost always seem plausible. Pre-thinking is what lets you cut through that confusion rather than flip-flopping under pressure.
There was a period where I was doing everything I was supposed to — practicing CR questions every day — but not seeing improvement. I brought this to my mentor Abha and she directed me back to specific course content: watching certain videos again, approaching the pre-thinking method more carefully. Not more questions, more method. That was a useful correction to make.
Reading Comprehension took longer. My instinct was to read every word carefully, understand every sentence before moving on. On GMAT RC with long passages and a clock running, that approach falls apart fast. The Master Comprehension module was what started to shift this — it reframed reading entirely. Not every word matters equally. Some passages are deliberately difficult and some of that difficulty is irrelevant to the questions. Your job is to understand the structure, not memorize the content.
The live session with Rajat sir was the turning point I hadn't expected. Watching him read a passage in real time — showing when to slow down, when to skim, how to use pause points to process structure rather than just absorb words — made something click that reading about the technique alone hadn't done. The pause points concept had been in my notes from the early lessons but seeing it applied line-by-line gave it a different kind of clarity. After that session I completely restructured how I approached RC. I stopped trying to understand everything and started reading for the architecture of the passage instead.
Cementing: Where the Real Improvement Happened
Scholaranium was the part of the platform I spent the most time in. The cementing quiz structure is set up deliberately — you start with relaxed time settings to focus on accuracy and process, then move to standard time once the method is working, then push into timed, harder practice.
I made the mistake early on of jumping to timed practice too fast. The accuracy wasn't there yet, and practicing under time pressure just reinforced the errors faster. When I pulled back to the relaxed setting and focused purely on process for two to three weeks, the accuracy numbers started moving. Only then did I tighten the time settings.
After each quiz, the PRISM Analysis showed exactly where I was strong and where I was bleeding. Not just "you got 60% on verbal" — it broke down by question type, by difficulty, by topic. That granularity mattered. It was the difference between knowing "I'm bad at CR" and knowing "I'm dropping points on weakening questions specifically when the argument involves a correlation-causation structure."
For quant, once the foundation modules were done, I used Scholaranium's custom quiz feature to target the specific sub-topics that kept showing up in my
error log. Probability and statistics were weak areas — not because I couldn't do the math, but because I was making the same conceptual errors repeatedly. I built custom quizzes filtered to just those topics, practiced them at increasing difficulty, and watched the
error log entries for those topics shrink week over week.
The
error log itself was tedious to maintain in the first couple of weeks. But by week four it was the most useful thing I had. You need the volume of questions to start seeing which mistakes are genuinely random and which ones are patterns. The log makes those patterns visible before they become habits.
Sectional Mocks, Full Mocks, and the Mental Game
Sectional mocks were where I validated each section before putting everything together. After completing verbal in Scholaranium, I'd run a sectional to confirm that the improvement was holding under actual exam conditions — timed, no pausing, no second chances. The sub-sectional breakdown after each sectional told me whether accuracy was holding across difficulty levels or just improving on medium questions.
When I moved to full-length mocks in the last few weeks, the focus shifted entirely to execution. Before each mock I'd use the first minute of each section to anchor: how many questions, how much time, how many I could afford to skip and revisit. This sounds basic but I had to build it as a habit. There were early mocks where I spent four or five minutes on a single question and didn't finish the section. Seeing that in the time distribution analysis after the mock made the problem impossible to ignore.
The mark-and-move strategy took practice to actually use. There's something psychologically uncomfortable about skipping a question you think you should be able to solve — especially as an engineer. I trained myself out of it through the sectionals first, deliberately practicing the skip, reviewing those questions afterward, realizing that the ones I came back to were usually more manageable than they'd seemed under initial pressure.
On test day, the second question in quant stopped me cold. Strong section, second question, got stuck. Old instinct said push through. What I actually did was estimate my remaining time, tried a different approach — working backwards from answer choices — and moved on. That adaptive response came from months of mock practice. Not from any single lesson, but from having been in that situation enough times to have a default.
The ISB Story
ISB's Round 3 deadline was January 25th — and their median score is 675.
I used AI to help structure my thinking — ISB professors have said openly that this is fine as long as your personal voice comes through. The framework I got from AI, I rewrote in my own language, with my own specific examples — programs I'd initiated, results I'd contributed to in my PSU role, things I hadn't fully appreciated until I sat down to articulate them.
The interview call came fast — February 14th, with my application submitted the last week of January. I had maybe ten days. But I realized I didn't need to prepare separately in the conventional sense. ISB's process is very much about understanding what you've written. Everything I'd already processed while writing the essays — why MBA, what I'd done, what I wanted to do, why ISB specifically — that became the interview preparation. The verbal work I'd put in had an unexpected bonus here too: the spot essays ISB assigns during the application process felt far less daunting after months of structured RC practice.
The panel told me in the interview itself that my clarity of goals stood out. That came from doing the application genuinely, not from coaching.
What I'd Tell Someone Starting Out
• Don't skip the foundation phase even when it feels slow. The concept → activity → practice file progression in e-GMAT exists for a reason. I was tempted multiple times to jump ahead into question practice before the method was solid. Every time I did, I ended up going back anyway. Build the foundation properly the first time.
• Review matters more than volume. The
error log felt tedious for the first two weeks. By week four it was the most useful thing I had. Practice without review just cements mistakes faster.
• Section sequencing isn't a preference, it's strategy. If you're tempted to rearrange the order because of comfort, think it through properly. I delayed that correction longer than I should have.
• The diagnostic is for honesty, not confidence. Go in without assumptions about what you know. I was a quant person who needed to go back and relearn number properties.
Attachments

1.png [ 79.63 KiB | Viewed 104 times ]